ScienceIDE is an independent infrastructure exploration for real scientific codebases, research tasks and verifiable scientific experience.
It turns real scientific code, research procedures, task goals and verification criteria into scientific environments that run, measure and reproduce, so that researchers can do agent execution, model training and reinforcement-learning research in real scientific settings.
Its value is in organising real scientific software, domain judgement, task goals and verifiable feedback into one executable environment, an experiential base for research agents, model training, reinforcement learning and discovery research. ScienceIDE does not require ScienceBuddy to be used, and does not depend on JEPA-Anything to stand.
Current task forms are accelerate, discover, repair, reproduce and integrate. ScienceIDE serves scientific code, scientific tasks, environment building, task verification and reinforcement-learning training on its own.
PhAI-IDE-72B, PhAI-IDE-9B and PhAI-IDE-4B are trained on interaction trajectories verified in these environments. The model family shows gains in held-out scientific-code repair and across selected general-purpose benchmarks in code, reasoning and knowledge: evidence of positive transfer from scientific experience to broader capabilities. These are observations on selected tasks and benchmarks, and how far the transfer reaches needs further evaluation.
The project page, the technical report and the weights for the three models are public, released by AItonomy Foundation on Hugging Face. The technical report has the full setup and results. ScienceIDE stays open for task proposals.